© 1995 American Heart Association, Inc.
Articles |
Synthetic Analogues of the Antithrombin III– Binding Pentasaccharide Sequence of Heparin
Prediction of In Vivo Residence Times
From the Scientific Development Group, N.V. Organon, Oss, the Netherlands.
Correspondence to R.G.M. van Amsterdam, Department of Vascular Pharmacology, N.V. Organon, PO Box 20, NL-5340 BH Oss, the Netherlands.
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResults and DiscussionReferences |
|---|
Abstract The synthetic pentasaccharide Org 31540/SR 90107A represents the antithrombin III (ATIII) binding region of heparin and accelerates the ATIII-mediated inhibition of coagulation factor Xa. This compound and 15 structural analogues with ATIII binding constants (Kd) ranging from 2.7 to 2600 nmol/L were compared for their plasma elimination in rats as measured from their factor Xa inhibiting activity. After administration of a low dose (100 nmol/kg body wt IV), each pentasaccharide showed a characteristic plasma half-life varying from a minimum of 0.3 hour for pentasaccharides with low affinity for ATIII to 10.9 hours for pentasaccharides with high affinity for the protein. The latter value was close to the half-life measured for radioiodinated rat ATIII (11.8 hours). We hypothesized that the elimination half-life of pentasaccharides is markedly extended by ATIII binding, of which the extent is governed by the Kd of the complex. The following observations support this hypothesis. The low-dose, low-affinity pentasaccharides were almost fully recovered in the urine without having lost anti–factor Xa activity, whereas compounds with high ATIII binding affinity were only partly recovered in the urine. With a high dose (500 nmol/kg body wt), a rapid plasma clearance of pentasaccharide was observed until a concentration similar to that of endogenous ATIII was reached, in accordance with their expected 1:1 stoichiometric interaction. The elimination half-life was similar to that of the low dose. The relation between Kd values and plasma half-lives could be explained by assuming rapid clearance of free and coclearance of ATIII-bound pentasaccharide with the protein. We applied the plasma ATIII concentration (3.5 µmol/L), the half-life of ATIII (11.8 hours), the half-life of unbound pentasaccharides (<10 minutes), and the Kd values and concluded that highly specific binding to ATIII in the circulation governs the presented straightforward pharmacological profile for the pentasaccharides.
Key Words: antithrombosis • pentasaccharides • in vivo residence • antithrombin III • rat
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResults and DiscussionReferences |
|---|
The anticoagulant heparin binds with high affinity to the plasma protein antithrombin III (ATIII) and changes its conformation. In this way heparin accelerates the ATIII-mediated inhibition of a number of serine proteases involved in coagulation. Until recently, only glycosaminoglycans of biological origin were available, which allowed investigators to derive only sparse structural information about this interaction. In the early 1980s, the unique pentasaccharide sequence representing the ATIII-binding region of heparin was described1 2 and produced by organic synthesis.3 4 5 6 This pentasaccharide catalyzes ATIII-mediated inactivation of coagulation factor Xa but not that of thrombin. An equieffective derivative containing a methoxy group at the reducing end (compound 15 in the Table
) was then
synthesized,5 and gradually more knowledge was gained
about the structural elements involved in ATIII
binding.7 8 9 10 11 12 13 14 15 A large series of synthetic analogues was
compared on the basis of structural differences within the individual
monosaccharide units, denoted D to H, starting from the nonreducing end
(Fig 1). Omission of specific charged groups within
these units leads to a dramatic decrease in ATIII binding. These groups
include O-sulfates at specific D,16
F,17 and G sites,18 the N-sulfate
at the F unit,19 and the carboxylate at
G.9 20 Addition of other groups leads to increased ATIII
binding10 11 compared with compound 15. For a specific
series of synthetic pentasaccharide analogues the kinetics of factor Xa
inactivation has recently been studied.21 Based on these
studies it was concluded that the difference in anti–factor Xa
activity among these pentasaccharides is exclusively due to differences
in ATIII binding affinity. Determination of the time courses of the
anti–factor Xa activity for the pentasaccharide analogues indicates
their duration of antithrombotic activity,22 23 24 and this
duration somehow correlates with the corresponding binding affinities
to ATIII.25 26 In the present study, the relation
between the plasma half-lives of pentasaccharides and their affinity
for ATIII was further explored, and a mechanism explaining the
observations is proposed.
|
|
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResults and DiscussionReferences |
|---|
Compounds
The pentasaccharides (Table
) were obtained by total organic
synthesis at N.V. Organon10 11 12 16 27 28 29 30 with research
collaboration with Sanofi Recherche, Gentilly, France.
ATIII Binding Affinities
The ATIII binding affinities of the synthetic
pentasaccharides were determined via measurements of inactivation of
factor Xa. Several concentrations of pentasaccharides were incubated at
pH 8.4 with 42 nmol/L human ATIII and 0.16 nkat/mL (2.4 nmol/L) bovine
factor Xa (final concentrations). After varying incubation times (0,
20, 40, or 60 seconds) at 37°C, a solution of chromogenic substrate
S-2765 (N-
-Cbo-D-Arg-Gly-Arg-pNA HCl;
Chromogenix AB) containing polybrene (hexadimethrine bromide; Sigma
Chemical Co) was added to obtain final concentrations of 0.2 mmol/L for
S-2765 and 0.5 mg/mL for polybrene. Thereafter the residual factor Xa
activity was determined by measuring the increase in absorbance at 405
nm over a fixed time interval by using a kinetic microtiter plate
reader. From these measurements the apparent first-order rate constant
of factor Xa inactivation was calculated as a function of the
pentasaccharide concentration, and from these dependencies the
dissociation constant (Kd) and second-order rate
constant of factor Xa inactivation were calculated according to Visser
et al21 for each of the pentasaccharides.
Ex Vivo Anti–Factor Xa Activities
Male Wistar Hsd/Cpb:Wu rats (body weight, 250 to 300 g) were
obtained from Harlan. One day prior to administration of the
pentasaccharides, the rats were anesthetized by injection of
methohexital sodium (40 g/L stock solution; 200 µL/100 g body wt IP;
Brietal, Eli Lilly). The right jugular vein was cannulated with a
siliconized, saline-filled PE-50 cannula (Clay Addams). The cannula was
led subcutaneously to the neck and exteriorized.
Some rats were nephrectomized by ligation of both kidneys. In contrast to the normal procedure, the cannulation of the jugular vein and subsequent administration of the pentasaccharides were performed immediately after the nephrectomy. The sampling period for these rats were restricted to a maximum of 22 hours.
Pentasaccharides were injected from stock solutions of 100 or 500 µmol/L in saline. Under light ether anesthesia, a volume of 100 µL/100 g body wt IV was administered via the penile vein, corresponding to a dose of 100 or 500 nmol/kg. Blood samples to a maximum of 500 µL were taken at preset time points from the cannula in plastic syringes containing 0.1 (vol/vol) sodium citrate (38 g/L ultrapure water) and centrifuged at 125 000 N/kg for 1 minute at room temperature. The supernatants were stored at -20°C until use.
Anti–factor Xa activities in rat plasma were measured amidolytically in an in vitro assay essentially according to Teien and Lie.31 All plasma samples were initially diluted fourfold with Tris-HCl buffer, pH 8.4, containing 50 mmol/L Tris, 100 mmol/L NaCl, and 7.5 mmol/L Na2-EDTA. For optimal calibration the samples were further diluted to a desired concentration range with fourfold diluted control rat plasma, which was used to ensure a constant ATIII level. Each diluted plasma sample was measured in duplicate in three different concentrations. Each 50-µL sample was supplemented with 50 µL human ATIII (0.25 U/mL; estimated 1 µmol/L) and 50 µL bovine factor Xa (0.75 nkat/mL; 11.3 nmol/L), both dissolved in the Tris-HCl buffer. After incubation for 2 minutes, 100 µL chromogenic substrate S-2222 (0.5 mmol/L Bz-Ile-Glu-Gly-Arg-pNA) was added, and absorbance was determined after 2 and 22 minutes at 405 nm. The anti–factor Xa activities were obtained from the differences in absorbance, subjected to logit transformation, and compared with a calibration curve. Compound 15 (Kd=754 nmol/L) was used to calibrate compounds with Kd values above 50 nmol/L and compound 7 (Kd=25.4 nmol/L) for compounds with Kd values equal to or below 50 nmol/L. The anti–factor Xa activity of compound 15 was determined against the fourth international standard of heparin and that of compound 7 against compound 15. Human ATIII, bovine factor Xa, and S-2222 were purchased from Kabi Vitrum.
Isolation and Labeling of Rat ATIII
ATIII was isolated at 4°C from 100 mL citrated rat plasma by
affinity chromatography on a heparin-Sepharose 6b (Pharmacia) column
(length, 300 mm; volume, 150 mL) at a flow rate of 60 mL/h. Unbound
protein was washed out by using a HEPES buffer, pH 7.4, containing 50
mmol/L HEPES, 100 mmol/L NaCl, and 0.1 mmol/L Na2-EDTA. For
gradient elution of ATIII, NaCl was added to the buffer to a maximum
concentration of 3 mol/L. ATIII-containing fractions, obtained between
1 and 1.6 mol/L NaCl, were combined and dialyzed against the HEPES
buffer, after which the gradient elution procedure was repeated. The
ATIII fraction was washed with saline and concentrated by using Amicon
Ym-10 ultrafiltration membranes (Mr 10 000
cutoff). This procedure yielded 5.5 mg protein in a volume of 6.5 mL.
By titration of anti–factor Xa activity with high-affinity
pentasaccharide (see below), an ATIII concentration of 6.5 µmol/L was
obtained, corresponding to a total of 2.5 mg protein and, thus,
biological viability was 0.46.
Part of the ATIII fraction, containing 5 µg protein in 0.05 mol/L phosphate buffer (pH 7.4), was radioiodinated by means of the lactoperoxidase method32 to a specific activity of 1.0x109 Bq/mg protein. Carrier-free [125I]NaI was obtained from NEN (Dupont Nemours GmbH; specific activity, 629x109 Bq/mg). A quantity of 0.16x106 Bq IV, corresponding to approximately 0.3 µg protein, was administered for determining the half-life of ATIII in individual rats. Electrophoresis and autoradiography of the ATIII fraction on a sodium dodecyl sulfate–polyacrylamide gel (0.07 wt/vol) before and after labeling did not show any measurable contamination by other proteins, and no biological loss in the ATIII activity of the labeled fraction was observed. The ATIII solution was stored at -80°C until use.
Estimation of Plasma ATIII Concentrations
The concentration of ATIII in rat plasma was estimated by two
separate approaches. The first estimate assumed a 1:1 stoichiometric
interaction between ATIII and pentasaccharides by amidolytical
measurement of anti–factor Xa activity in diluted plasma in the
presence of a pentasaccharide with very high binding affinity to ATIII
(Kd<5 nmol/L). Starting from an
ATIII-saturating pentasaccharide concentration, stepwise lower
concentrations of pentasaccharide were added to the diluted plasma.
Thereafter, factor Xa was added, and exactly 2 minutes later
inactivation of factor Xa was quenched by addition of polybrene to a
final concentration of 0.5 mg/mL. The residual factor Xa activities
were determined by using S-2222 as described above to establish the
lowest ATIII-saturating concentration, which is considered to equal the
ATIII concentration. Below that concentration an increasing part of the
ATIII is left unbound and no longer contributes significantly to the
anti–factor Xa activity. This is reflected by a proportionally higher
residual factor Xa activity (Fig 2). Rat and human
plasmas were prediluted 60 times in Tris-HCl buffer, pH 8.4, to ensure
that in the absence of pentasaccharide no anti–factor Xa activity was
detected, and in its presence inhibition of factor Xa (initial
concentration, 10 nmol/L) would be proportionally related to the size
of the pentasaccharide-bound fraction of ATIII. The purified rat ATIII
was quantified by this method after 75 times predilution of the stock
with Tris-HCl buffer, pH 8.4, supplemented with 1% polyethylene
glycol–6000.
|
) and human (
) plasmas in the presence of
decreasing concentrations of a pentasaccharide (penta) with high
antithrombin III (ATIII) binding affinity (Kd<5
nmol/L). Arrows pointing to the solid symbols indicate the minimum
ATIII-saturating concentrations of pentasaccharide that reflect the
plasma concentration of ATIII. Data represent mean±SEM of
three separate experiments.
A second estimate ([ATIII]0) was based on
Equations 1 and 2, which relate the first-order rate constant of factor
Xa inactivation, determined at pseudo first-order rate conditions
(kobs), to the given increasing initial
pentasaccharide concentration ([P]0) for a
high-affinity pentasaccharide in the mixture of free and complexed
pentasaccharide with differently diluted (10- to 60-fold) ATIII rat
plasmas.
 | (1) |
 | (2) |
Determination of Plasma Pentasaccharide Concentrations Not Bound to
ATIII
Rats were administered pentasaccharides at a dose of 500 nmol/kg
to give a plasma concentration exceeding the concentration required to
saturate circulating ATIII. Volumes of 150 µL plasma obtained from
consecutive blood samples (500 µL) were centrifuged for 30 minutes at
20 000 N/kg) through Millipore filters (Ultrafree CL) with an
Mr 30 000 cutoff to exclude passage of ATIII.
The anti–factor Xa activity of the plasma filtrate was determined
after reconstitution with human ATIII as described above. This activity
was considered to be proportional to the plasma pentasaccharide
fraction, which is unbound to ATIII. The percent free of the total
amount of free and bound pentasaccharide was calculated from the ratio
of anti–factor Xa activities in filtrated and nonfiltrated plasma. The
mean value was taken of the last four observations in the elimination
phase.
The percent unbound pentasaccharide was also predicted by calculation
using Equation 2. After administration of pentasaccharide, the time
courses of the ATIII-pentasaccharide complex concentrations are
reflected by the total anti–factor Xa activity, and
[ATIII]0 is assumed to be constant (3.5
µmol/L). [P]0 was calculated for all time
points of blood sampling.
Data Analysis
Plasma half-lives of anti–factor Xa activity wee computed by
using MW \ PHARM software (Medi \ ware), which uses the
Simplex method, in which the equation for biexponential disappearance
(Equation 4) is fitted to the plasma activities obtained. Subsequently,
the elimination half-lives were calculated by using weighting factors
that compensated for the relative and absolute error, the latter being
independent of the concentration. Bioavailabilities were calculated as
areas under the curve (AUC) obtained from the fitted polyexponential
curves extrapolated to infinity. Volumes of distribution
(Vd) in milliliters per kilogram were computed (using
MW \ PHARM) as the quotient of doses (D) and
initial plasma anti–factor Xa activities expressed in micromoles per
liter.
 | (3) |
, which describes the biexponential
disappearance of a pentasaccharide at a concentration [P]
and a given time point t.
 | (4) |
and ß. All data are expressed as mean±SEM. The statistical significance of differences was estimated by using Student's unpaired t test. Nonlinear least-squares regression analysis was performed according to the Marquardt method using the NLIN SAS procedure. The level of significance was set at P=.05.
|
Results and Discussion |
|---|
|
TopAbstractIntroductionMethodsResults and DiscussionReferences |
|---|
Sixteen structural analogues of the ATIII binding pentasaccharide were prepared by organic synthesis and tested in vitro and in vivo (Table
). The dissociation constants (Kd values)
for ATIII binding ranged from 2.7 to 2600 nmol/L, the plasma
elimination half-lives from 0.3 to 10.9 hours, and the
bioavailabilities (AUC values) from 0.3 to 13.0
h · µmol · L-1. In general, with increasing
Kd values, decreasing half-lives and
bioavailabilities were observed. Compound 5 had the longest half-life
measured (10.9 hours). This half-life approached the 11.8±0.6 hours
(n=4) obtained for circulating ATIII using purified radioiodinated rat
ATIII. Pentasaccharides with very low binding affinities for ATIII
displayed very high clearance rates, which made it difficult to
distinguish elimination from distribution. Fig 3
illustrates the elimination of 125I-ATIII and for a series
of 100 nmol/kg IV pentasaccharides.
|
) and of the anti–factor Xa activity of the
pentasaccharides referred to in the Table
), 7
(
), 12 (
) after injection of
100 nmol/kg IV in rats. Values are expressed as percent of the first
data point obtained 1 minute after injection and represent
means of three separate experiments.
The Hypothesis
On the basis of the previous observations, we hypothesized that
the elimination half-life of pentasaccharides is markedly extended by
selective ATIII binding, which is dependent on the
Kd of the complex. Equation 5 reflects these
observations as it describes the relation of the plasma half-life
(T1/2) and the Kd values
of pentasaccharides, in which Kec
represents the elimination rate constant of the complex, which
is assumed to equal that of circulating ATIII, and
Kef represents the elimination rate
constant of free pentasaccharide.
 | (5) |
0, the half-life of measured anti–factor
Xa activity (T1/2) equals
ln2/Kec, implying that among
pentasaccharides with increasing binding affinity the half-life of the
administered compound approaches that of ATIII. In the reverse case, if
a given pentasaccharide does not bind ATIII at all, ie,
Kd
, then
T1/2=ln2/Kef. This
implies that with decreasing ATIII binding affinity the measured
half-life approaches that of free pentasaccharide.
According to this hypothesis and supported by the experimental data, a
sigmoidal relation exists between the dissociation constants of the
various pentasaccharide-ATIII complexes and the corresponding plasma
half-lives of the individual pentasaccharides. This relation, as
described by Equation 5 with a Kec of 0.059
h-1 (T1/2=11.8 hours) and a plasma
ATIII concentration of 3.5 µmol/L, was fitted to the data set, and a
Kef was obtained of 5.78 h-1
(T1/2=7.2 minutes). Fig 4 (top)
shows two curves, the solid line representing the curve
obtained if ATIII-bound pentasaccharides are cleared as complexes and
the dashed line representing the other extreme situation, if
complexes are not cleared (Kec=0); ATIII would
thus only be eliminated after dissociation from the complex
(Kef=13.9 h-1;
T1/2=3 minutes). The latter would imply that
pentasaccharide binding delays the clearance of ATIII, which contrasts
with experimental evidence. In these experiments rat
125I-ATIII was presaturated with compound 5
(Kd=13.1 nmol/L) and administered to rats
receiving an infusion of that pentasaccharide at a dose that
continuously saturated the ATIII in circulation. A half-life of
12.0±0.1 hours (n=3) was obtained for 125I-ATIII,
indicating that the half-life of ATIII is not significantly affected by
pentasaccharide binding. This lack of effect on elimination of ATIII by
a pentasaccharide contrasts with the decrease observed upon heparin
binding.34 A second argument against separate clearance of
pentasaccharides and ATIII, and in favor of clearance of
pentasaccharide-ATIII complex as a whole, is that with increasing ATIII
binding affinity a maximum elimination half-life is obtained that does
not exceed that of ATIII. This is best illustrated by Fig 4 (bottom),
which shows the relation between the association constants
(Ka=1/Kd) and the plasma
elimination half-lives of the individual pentasaccharides.
|
It is to be emphasized that Kd and Ka values were applied as obtained using human ATIII at pH 8.4. In preliminary studies with rat ATIII at a physiological pH two to three times stronger, binding affinity for a number of tested pentasaccharides was observed. This implies a parallel shift of the curve to the left and is unlikely to change the conclusions drawn.
Role of the Kidneys
To explain the delayed clearance of pentasaccharides by ATIII
binding, a large difference must exist between the elimination rate of
free and ATIII-bound pentasaccharides. Moreover, this difference is
assumed to be similar for all pentasaccharides. Since the kidneys are
known to play a major role in the clearance of
heparins,35 36 37 38 39 40 it was anticipated that the rapid clearance
of unbound pentasaccharide would be associated with renal filtration.
Accordingly, after intravenous administration compound 15 showed a
24-hour urine recovery of anti–factor Xa activity of 89±2% (n=3).
Accumulation in the urine started shortly after plasma clearance had
begun (Fig 5). For pentasaccharides with higher
affinities, partial recoveries were obtained of 52±2% for compound 12
(Kd=135 nmol/L) over a 24-hour period and
37±6% for compound 5 (Kd=13.1 nmol/L) over 72
hours. These findings agree with the idea that ATIII binding protects
pentasaccharides from renal clearance, and thus alternative routes of
elimination become important, possibly associated with the clearance of
ATIII. For drugs that are fully renally cleared, allometry can be used
to predict from small-animal data elimination half-lives in larger
animals and humans. This has also been applied for pentasaccharides, in
which a half-life of 96 hours has been predicted for compound 7 in
humans,41 whereas the model presented here would
estimate between 50 and 60 hours. Ceustermans et al,42
after comparing the half-life of covalently linked heparin-ATIII
complexes with that of free administered heparin, concluded that
clearance of the anticoagulant activity of heparin is preceded by
dissociation of the heparin-ATIII complex instead of clearance of the
complex as a whole. This does not conflict with our observations
because the weak ATIII binding affinity of the pentasaccharide sequence
in heparin permits a complete urine recovery, as shown for compound 15
having a comparable ATIII binding affinity.
|
Role of the Liver
Little specific information is available concerning the route of
elimination of ATIII. Mechanisms have been described for the
elimination of ATIII bound to a target serine protease. For example,
ATIII-thrombin complexes are rapidly cleared by internalization in the
liver after specific binding to a hepatocyte receptor.39
And ATIII that is not bound to a target serine protease can be cleared
by the liver. Glycoproteins in circulation like ATIII are exposed to a
neuraminidase that randomly removes terminal sialic acid residues from
the carbohydrate chains attached to the protein surface. The
glycoprotein is then changed into an asialoglycoprotein with selective
high affinity to a liver receptor referred to as (rat) hepatic lectin
(for review, see Reference 4343 ). Receptor-bound asialoglycoproteins are
considered to be subjected to endocytosis via coated pits and vesicles,
transferred to lysozomes, and degraded. That a fraction of the
pentasaccharide-ATIII complexes is not dissociated during clearance is
suggested by the observed reduction of urine recovery for high-affinity
pentasaccharides. That ATIII is not renally secreted is concluded from
the observation that after nephrectomy a half-life of rat
125I-ATIII was obtained of 11.7±0.5 hours (n=4), not
different from the 11.8 hours obtained with controls.
Plasma Concentration of ATIII
The plasma concentration of ATIII in rats was assessed in
titration experiments that monitored the dose-dependent decrease of
residual factor Xa activity for a pentasaccharide with high affinity
for ATIII. A value of 2.8 µmol/L (n=3) was consistently found for rat
plasma; the ATIII concentration in the human plasma that was included
for reference amounted to 3.2±0.2 µmol/L (n=4). The other method,
using the measured first-order rate constant as a parameter, provided
an ATIII concentration in rat plasma of 4.2 µmol/L and in human
plasma of 3.5 µmol/L. The mean value for rat plasma obtained from the
two approaches, 3.5 µmol/L, was used for further calculations. The
values for human plasma are within published ranges.44 45 46 47 48
Conard et al45 have indicated that plasma ATIII levels can
be underestimated when obtained by the active-site titration technique
with thrombin in the presence of heparin because heparin can interfere
with the stoichiometry of thrombin binding by ATIII.49 Due
to its low molecular weight, a pentasaccharide is less likely to modify
the presently applied titration for factor X activation.
Clearance of Excess Pentasaccharide Compared With ATIII
An initial distribution volume (compartment A) of 65±3
mL/kg was computed as a mean value for all pentasaccharides
administered at 100 nmol/kg. A dose of approximately 230 nmol/kg IV was
calculated to give a plasma concentration of 3.5 µmol/L,
corresponding with that of ATIII. On the basis of that calculation, 500
nmol/kg of the pentasaccharides was administered to give a molar plasma
concentration exceeding that of circulating ATIII. The administered
surplus circulates unbound to ATIII in the plasma and is cleared in the
distribution phase of the time curve (Fig 6). At that
high dose some pentasaccharides were found to have a larger volume of
distribution than at 100 nmol/kg, ranging from 67 to 142 mL/kg. These
volumes correspond with plasma concentrations of 7.5 to 3.4 µmol/L
(4.9±0.4 µmol/L) at time zero. Thus, whereas the distribution volume
of low doses is restricted to that of ATIII, ie, the blood plasma,
excess of unbound pentasaccharides may either cross the vessel wall to
enter the interstitial space or be cleared instantaneously after
administration. Despite the increased distribution volume for a dose of
500 nmol/kg, the obtained plasma concentrations still exceeded that of
circulating ATIII. The clearance of the surplus pentasaccharide
compared with ATIII was confirmed to be rapid, as plasma concentrations
decreased within 5 minutes to 3.2±0.2 µmol/L, the level of
circulating ATIII. For four of the five pentasaccharides with the
lowest Kd values, only a minor further decrease
of the plasma anti–factor Xa activity was observed in the 25 minutes
thereafter to a level corresponding with a plasma concentration of
2.7±0.2 µmol/L. This reflects a plasma retention of pentasaccharides
by binding to ATIII.
|
, 
, 
,
Clearance of Unbound Pentasaccharide During Steady-State
Equilibrium for ATIII Binding
After elimination of the administered surplus of pentasaccharide
over ATIII, unbound pentasaccharide is rapidly cleared from the
circulation. However, this clearance is restricted by the percent free
in circulation, which is governed by the association-dissociation
equilibrium of pentasaccharide-ATIII complexes, which in turn is
determined by the Kd values. As a consequence,
after removal of the excess amount only, which is not bound to ATIII in
circulation, dose-independent elimination half-lives of
pentasaccharides were found. This is illustrated by Fig 6, which shows
the parallel time courses of plasma anti–factor Xa activity for 500
versus 100 nmol/kg of four different pentasaccharides expressed as
plasma concentrations. For heparin, in contrast, the plasma half-life
for different doses is differently affected by saturable endothelial
binding sites.50 51 Pentasaccharides do not bind to
endothelial cells, and for low-molecular-weight (LMW) heparins the
degree of binding depends on the molecular weight and degree of
sulfation.52 The pharmacokinetic behavior of heparin is
also complicated by the binding to numerous plasma
proteins53 and metabolism by endoglycosidases and
endosulfatases within the reticuloendothelial system.54 55 56 57
Although pentasaccharides showed functional stability in blood plasma,
it cannot be excluded that some of the sulfate groups are eliminated in
this metabolic process since functionality is not always measurably
affected after desulfation. That the pharmacokinetic properties of LMW
heparins can be positioned between heparin and pentasaccharides is
indicated by observations that the LMW heparins tend to show a less
complex, ie, more linear, elimination than heparin.58 59
However, the pharmacokinetic behavior of LMW heparins still does not
allow analysis as performed here, mainly because of their chemical
heterogeneity.
After we established that the elimination of pentasaccharides after
distribution and clearance of the administered surplus is delayed by
ATIII binding, experiments were performed to verify whether the plasma
concentrations of free and ATIII-bound fractions of administered
pentasaccharides are accurately predicted from the
Kd values. After administration of
pentasaccharides at a dose of 500 nmol/kg, plasma samples were taken,
and the unbound fraction was isolated by ultrafiltration from the
individual samples. Time courses of total plasma anti–factor Xa
activity and the activity of the unbound pentasaccharide fraction (Fig 7), expressed as plasma concentrations, show a very
rapid initial clearance of the unbound fraction to a plasma level five
times less than that of the total activity for compound 15
(Kd=754 nmol/L) and 60 times less than that of
the total activity for compound 9 (Kd=31.1
nmol/L). After more than 90% of the pentasaccharide was cleared, the
total/free ratios became constant because further proportions of the
two fractions were eliminated equally rapidly as reflected by the
parallel semi-log curves. This confirms that the percent of
pentasaccharide that circulates unbound to ATIII is then fully
determined by the dissociation constant of the complex, and renal
clearance is no longer rate limiting. In addition, the dotted lines
without symbols in Fig 7 show the predicted time courses of plasma
concentrations for the two examples of unbound pentasaccharides as
calculated for each individual time point using Equation 2. These
curves match very well with those empirically obtained (open symbols),
showing that association-dissociation equilibria that are obtained in
vitro and calculated with the given mathematical model are also found
in vivo despite the presence of potentially interfering proteins or
cells.
|
Exceptions to the Rule
For one pentasaccharide, compound 13
(Kd=162 nmol/L), a plasma half-life of 0.5 hour
was observed, which is much smaller than expected from its ATIII
binding affinity and also from the measured fraction of unbound
pentasaccharide in the plasma (3%). Moreover, the 24-hour urine
recovery, which was expected by its half-life to be of the same order
of magnitude as that of compound 15 (89), amounted to only 36±1%
(n=3). These data clearly indicate that this pentasaccharide is either
subjected to an alternative route of plasma clearance that is not
associated with the renal route or clearance of ATIII or to metabolic
breakdown. This was confirmed by using nephrectomized rats, in which
the 0.7-hour plasma half-life of compound 15 was increased to 14 hours,
but that of compound 13 was increased from 0.5 hour to 2 hours. Another
pentasaccharide, compound 1, showed a more rapid clearance than
expected from its high ATIII binding affinity
(Kd=2.7 nmol/L). Detailed comparison of the
structures of all tested compounds suggests that a high degree of
sulfation in the D unit may be involved. For this pentasaccharide it
was also noticed that the 3% unbound to ATIII as obtained from plasma
filtrations was in the same order as for compound 7
(Ka=25.4 nmol/L; 4% unbound) and much higher
than the 0.1% calculated from its Kd value.
This might indicate that the assay used to determine the
Kd values can lead to overestimation of the
ATIII binding affinity or, alternatively (for this particular
pentasaccharide only), the Kd value for rat
ATIII is higher than for human ATIII measured at pH 8.4.
In conclusion, straightforward pharmacokinetic properties are presented for the majority (14 of 16) of pentasaccharides representing a new class of fully synthetic antithrombotic compounds. In contrast to the complex pharmacokinetics of heparin, three elements are distinguished that determine the overall in vivo residence time of pentasaccharides: the turnover rate of circulating ATIII, the renal clearance rate of unbound pentasaccharide, and the dissociation constant (Kd) for the circulating pentasaccharide-ATIII complexes. With elimination half-lives similar to that of ATIII and only partial overall recovery in the urine (<40%), high-affinity pentasaccharides are most likely to be cleared together with ATIII, possibly by the liver. The proposed mechanism provides a means to predict time courses of plasma anti–factor Xa activities for pentasaccharides in humans from their Kd value, the elimination half-life of 66 hours60 for ATIII, the plasma ATIII concentration, and the half-life of any one unbound pentasaccharide. These plasma anti–factor Xa activities are highly correlated with the antithrombotic efficacy of these drugs.
|
Acknowledgments |
|---|
Drs C. van Boeckel and P. Westerduin of our Department of Medicinal Chemistry III are acknowledged for constructive discussions and Dr M. Smit, Department of Vascular Pharmacology, for intensive reviewing of the manuscript.
Received August 26, 1994; accepted January 18, 1995.
|
References |
|---|
|
TopAbstractIntroductionMethodsResults and DiscussionReferences |
|---|
1. Choay J, Lormeau JC, Petitou M, Sinay P, Fareed J. Structural studies on a biologically active hexasaccharide obtained from heparin. Ann N Y Acad Sci. 1981;370:644-649. [Medline] [Order article via Infotrieve]
2. Thunberg L, Bäckström G, Lindahl U. Further characterization of the antithrombin-binding sequence in heparin. Carbohydr Res. 1982;100:393-410. [Medline] [Order article via Infotrieve]
3. Choay J, Petitou M, Lormeau JC, Sinay P, Casu B, Gatti G. Structure-activity relationship in heparin: a synthetic pentasaccharide with high affinity for antithrombin III and eliciting high anti-factor Xa activity. Biochem Biophys Res Commun. 1983;116:492-499. [Medline] [Order article via Infotrieve]
4. Sinay P, Jacquinet JC, Petitou M, Duchaussoy P, Lederman I, Choay J, Torri G. Total synthesis of a heparin pentasaccharide fragment having high affinity for antithrombin III. Carbohydr Res. 1984;132:C5-C9.
5. Van Boeckel CAA, Beetz T, Vos JN, De Jong AJM, Van Aelst SF, Van der Bosch RH, Mertens JMR, Van der Vlugt FA. Synthesis of pentasaccharide corresponding to the antithrombin III binding fragment of heparin. J Carbohydr Chem. 1985;4:293-321.
6. Ichikawa Y, Monden R, Kuzuhara H. Synthesis of a heparin pentasaccharide fragment with high affinity for antithrombin III employing cellobiose as a key starting material. Tetrahedron Lett. 1986;27:611-614.
7.
Löbermann H, Tokuoka R, Deisenhofer J, Huber R. Human
1-proteinase inhibitor: crystal structure analysis
of two crystal modifications, molecular model and preliminary
analysis of the implications for function. J Mol Biol. 1984;177:531-556. [Medline]
[Order article via Infotrieve]
8.
Huber R, Carrell RW. Implications of the three dimensional
structure of 1-antitrypsin for the structure and
function of serpins. Biochemistry. 1989;28:8952-8966.
9.
Van Boeckel CAA, Lucas H, Van den Nieuwenhof MWP, Wagenaars
GN, Mellema JR. Synthesis and conformational analysis of an
analogue of the antithrombin-binding region of heparin: the role of the
carboxylate function of -L-idopyranuronate. Recl Trav Chim
Pays Bas. 1987;106:581-591.
10. Van Boeckel CAA, Van Aelst SF, Beetz T, Meuleman DG, van Dinther TG, Moelker HCT. Structure-activity relationships of synthetic heparin fragments: discovery of a very potent AT-III activating pentasaccharide. Ann N Y Acad Sci. 1989;556:489-491.
11. Van Boeckel CAA, Beetz T, Van Aelst SF. Synthesis of a potent antithrombin activating pentasaccharide: a new heparin-like fragment containing two 3-O-sulphated glucosamines. Tetrahedron Lett. 1988;29:803-806.
12. Beetz T, Van Boeckel CAA. Synthesis of an antithrombin binding heparin like pentasaccharide lacking 6-O-sulfate at its reducing end. Tetrahedron Lett. 1986;27:5889-5892.
13. Grootenhuis PKJ, Van Boeckel CAA. Constructing a molecular model of the interaction between antithrombin III and a potent heparin analogue. J Am Chem Soc. 1991;113:2743-2747.
14. Petitou M, Lormeau JC, Choay J. A new synthetic pentasaccharide with increased anti-factor Xa activity: possible role for anionic clusters in the interaction of heparin and antithrombin III. Semin Thromb Hemost. 1991;17(suppl 2):143-146.
15. Jaurand G, Basten J, Lederman I, Van Boeckel CAA, Petitou M. Biologically active heparin-like fragments with a `non-glycosamino'glycan structure, I: a pentasaccharide containing a 3-O-methyl iduronic acid unit. Bioorg Med Chem Lett. 1992;2:897-900.
16.
Lindahl U, Bäckström L, Thunberg L. The
antithrombin binding sequence in heparin: identification of an
essential 6-O-sulfate group. J Biol Chem. 1983;258:9826-9830.
17.
Lindahl U, Bäckström L, Thunberg L, Leder IG.
Evidence for a 3-O-sulfated D-glucosamine residue in the
antithrombin-binding sequence of heparin. Proc Natl Acad Sci
U S A. 1980;77:6551-6555.
18. Petitou M, Lormeau JC, Choay J. AT III binding site of heparin: determination of the role of sulfate groups at the reducing-end disaccharide through new synthetic pentasaccharides. Thromb Haemost. 1987;58:7. Abstract.
19.
Riesenfeld J. Thunberg L, Höök M, Lindahl U. The
antithrombin-binding sequence of heparin: location of essential
N-sulfate groups. J Biol Chem. 1981;256:2389-2394.
20.
Van Aelst SF, Van Boeckel CAA. Synthesis of an analogue
of the antithrombin binding region of heparin containing
-L-idopyranose. Recl Trav Chim Pays Bas. 1987;106:593-595.
21. Visser A, Buiting MT, Van Dinther TG, Van Boeckel CAA, Grootenhuis PG, Meuleman DG. The ATIII-binding affinities of a series of synthetic pentasaccharide analogues. Thromb Haemost. 1991;65:1296. Abstract.
22. Hobbelen PMJ, Van Dinther TG, Vogel GMT, Van Boeckel CAA, Moelker HCT, Meuleman DG. Pharmacological profile of the chemically synthesized antithrombin III binding fragment of heparin (pentasaccharide) in rats. Thromb Haemost. 1990;63:265-270. [Medline] [Order article via Infotrieve]
23. Meuleman DG, Hobbelen PMJ, Van Dinther TG, Vogel GMT, Van Boeckel CAA, Moelker HCT. Antifactor Xa activity and antithrombotic activity in rats of structural analogues of the minimum antithrombin III binding sequence: discovery of compounds with longer duration of action than the natural pentasaccharide. Semin Thromb Hemost. 1991;17(suppl 1):112-117.
24. Walenga JM, Fareed J, Petitou M, Samama M, Lormeau JC, Choay J. Intravenous antithrombotic activity of a synthetic heparin pentasaccharide in a human serum induced stasis thrombosis model. Thromb Res. 1986;43:243-248. [Medline] [Order article via Infotrieve]
25. Vogel GMT, Van Dinther TG, Visser A, Buiting MT, Van Amsterdam RGM, Meuleman DG. Significance of the ATIII binding affinity of some pentasaccharide analogues for their anti Xa activities and elimination rates in rats. Thromb Haemost. 1991;65:930. Abstract.
26. Van Amsterdam RGM, Vogel GMT, Visser A, Kop W, Buiting MT, Meuleman DG. Plasma disappearance of synthetic pentasaccharides derived from heparin explained by antithrombin-III binding. Thromb Haemost. 1993;69:893. Abstract.
27. Basten J, Jaurand G, Olde-Hanter B, Petitou M, Van Boeckel CAA. Biologically active heparin-like fragments with a `non-glycosamino'glycan structure, II: tetra-O-methylated pentasaccharide with high affinity for antithrombin III. Bioorg Med Chem Lett. 1992a;2:901-904.
28. Basten J, Jaurand G, Olde-Hanter B, Duchaussoy P, Petitou M, Van Boeckel CAA. Biologically active heparin-like fragments with a `non-glycosamino'glycan structure, III: O-alkylated-O-sulphated pentasaccharides. Bioorg Med Chem Lett. 1992b;2:905-910.
29. Petitou M, Van Boeckel CAA. Chemical synthesis of heparin fragments and analogues. In: Herz W, Kirby GW, Tamm Ch, eds. Progress in the Chemistry of Organic Natural Products. Vienna, Austria: Springer Verlag; 1992:144–210.
30. Lucas H, Basten JEM, Konradsson P, Van Boeckel CAA. A short synthetic route towards a biologically active heparin-like pentasaccharide displaying a pseudo alternating sequence. Angew Chemie Int Ed. 1993;32:434-436.
31. Teien AN, Lie M. Evaluation of an amidolytic heparin assay method: increased sensitivity by adding purified antithrombin III. Thromb Res. 1977;10:399-410. [Medline] [Order article via Infotrieve]
32. Marchalonis JJ. An enzymatic method for the trace iodination of immunoglobulins and other proteins. Biochem J. 1969;113:299-305. [Medline] [Order article via Infotrieve]
33. Sheiner LB. ELSFIT Users Manual. San Francisco, Calif: Division of Clinical Pharmacology of the University of California; 1983. Technical report.
34. De Swart CAM, Nijmeijer B, Andersson IO, Holmer E, Sixma JJ, Bouma BN. Elimination of intravenously administered radiolabelled antithrombin III and heparin in humans. Thromb Haemost. 1984;52:66-70. [Medline] [Order article via Infotrieve]
35. Danishefsky I, Eiber HB. Studies on the metabolism of heparin. Arch Biochem Biophys. 1959;85:53-61. [Medline] [Order article via Infotrieve]
36. Shanberge JN, Gruhl M, Kitani T, Ambegaonkar S, Kambayashi J, Nakagawa M, Lenter D. Fractionated tritium-labelled heparin studied in vitro and in vivo. Thromb Res. 1987;13:767-783.
37. Caranobe C, Barret A, Gabaig AM, Dupouy D, Sié P, Boneu B. Disappearance of circulating anti-Xa activity after intravenous injection of standard heparin and a low molecular weight heparin (CY 216) in normal and nephrectomized rabbits. Thromb Res. 1985;40:129-133. [Medline] [Order article via Infotrieve]
38. Boneu B, Caronobe C, Gabaig AM, Dupouy D, Sie P, Buchanan MR, Hirsh J. The disappearance of a low molecular weight heparin fraction (CY 216) differs from standard heparin in rabbits. Thromb Res. 1987;46:845-853. [Medline] [Order article via Infotrieve]
39.
Shifman MA, Pizzo SV. The in vivo metabolism of
antithrombin III and antithrombin III complexes. J Biol
Chem. 1982;257:3243-3248.
40. Palm M, Mattsson C. Pharmacokinetics of heparin and low molecular weight heparin fragment (fragmin®) in rabbits with impaired renal and metabolic clearance. Thromb Haemost. 1987;58:932-935. [Medline] [Order article via Infotrieve]
41. Crepon B, Donat F, Bârzu T, Hérault JP. Pharmacokinetic (PK) parameters of ATIII binding pentasaccharides in three animal species: predictive value for humans. Thromb Haemost. 1993;69:654. Abstract.
42.
Ceustermans R, Hoylaerts M, De Mol M, Collen D. Preparation,
characterization, and turnover properties of heparin-antithrombin III
complexes stabilized by covalent bonds. J Biol Chem. 1982;257:3401-3408.
43.
Drickamer K. Two distinct classes of carbohydrate-recognition
domains in animal lectins. J Biol Chem. 1988;263:9557-9560.
44. Murano G, Williams L, Miller-Andersson M, Aronson DL, King C. Some properties of antithrombin-III and its concentration in human plasma. Thromb Res. 1989;18:259-262.
45. Conard J, Brosstad F, Larsen ML, Samama M, Abildgaard U. Molar antithrombin concentration in normal human plasma. Haemostasis. 1983;13:363-368. [Medline] [Order article via Infotrieve]
46. Christensen U. Requirements for valid assays of clotting enzymes using chromogenic substrates. Thromb Haemost. 1980;43:169-174. [Medline] [Order article via Infotrieve]
47. Collen D, Schetz J, De Cock F, Holmer E, Verstraete M. Metabolism of antithrombin III (heparin cofactor) in man: effects of venous thrombosis and of heparin administration. Eur J Clin Invest. 1977;7:27-35. [Medline] [Order article via Infotrieve]
48. Chan VTK, Wong V, Tso SC, Todd D. The determination of antithrombin III by radioimmunoassay and its clinical application. Br J Haematol. 1979;41:563-572. [Medline] [Order article via Infotrieve]
49. Hoylaerts M, Holmer M, De Mol M, Collen D. Covalent complexes between low molecular weight heparin fragments and antithrombin III: inhibition kinetics and turnover parameters. Thromb Haemost. 1983;49:109-115. [Medline] [Order article via Infotrieve]
50. Mahadoo J, Hiebert L, Jaques LB. Vascular sequestration of heparin. Thromb Res. 1977;12:79-90.
51. Glimelius B, Busch C, Höök M. Binding of heparin on the surface of cultured human endothelial cells. Thromb Res. 1978;12:773-782. [Medline] [Order article via Infotrieve]
52. Bârzu T, Van Rijn JML, Petitou M, Molho P, Tobelem G, Caen JP. Endothelial binding sites for heparin: specificity and role in heparin neutralization. Biochem J. 1986;238:847-854. [Medline] [Order article via Infotrieve]
53. Marciniak E. Binding of heparin in vitro and in vivo to plasma proteins. J Lab Clin Med. 1974;84:344-356. [Medline] [Order article via Infotrieve]
54. Dietrich CP. A heparin sulfamidase from mammalian lymphoid tissue. Can J Biochem. 1970;48:725-733. [Medline] [Order article via Infotrieve]
55. Friedman Y, Arsenis C. Studies on the heparin sulphamidase activity from rat spleen: intracellular distribution and characterization of the enzyme. Biochem J. 1974;139:699-708. [Medline] [Order article via Infotrieve]
56.
Höök M, Pettersson I,
ren S. A
heparin-degrading endoglycosidase from rat spleen. Thromb
Res. 1977;10:857-861. [Medline]
[Order article via Infotrieve]
57. Fabian I, Bleiberg I, Aronson M. Increased uptake and desulphation of heparin by mouse macrophages in the presence of polycations. Biochim Biophys Acta. 1978;544:69-76. [Medline] [Order article via Infotrieve]
58. Harenberg J, Gnasso A, de Vries J, Zimmermann R, Augustin I. Anticoagulant and lipolytic effects of a low molecular weight heparin fraction. Thromb Res. 1985;39:683-692. [Medline] [Order article via Infotrieve]
59. Bratt G, Törnebohm E, Lockner D, Bergström K. A human pharmacological study comparing conventional heparin and a low molecular weight heparin fragment. Thromb Haemost. 1985;53:208-211. [Medline] [Order article via Infotrieve]
60. Knot EAR, De Jong E, Ten Cate JW, Gie LK, Van Royen EA. Antithrombin III: biodistribution in healthy volunteers. Thromb Haemost. 1987;58:1008-1011.[Medline] [Order article via Infotrieve]
This article has been cited by other articles:
 |
|
 |
|
  K. Yamashiro, J. Kiryu, A. Tsujikawa, M. Honjo, A. Nonaka, K. Miyamoto, Y. Honda, H. Tanihara, and Y. Ogura Inhibitory Effects of Antithrombin III against Leukocyte Rolling and Infiltration during Endotoxin-Induced Uveitis in Rats Invest. Ophthalmol. Vis. Sci., June 1, 2001; 42(7): 1553 - 1560. [Abstract] [Full Text]
|
|
|
|
 |
|
 J.M. Herbert, J.P. Herault, A. Bernat, R.G.M. van Amsterdam, J.C. Lormeau, M. Petitou, C. van Boeckel, P. Hoffmann, and D.G. Meuleman Biochemical and Pharmacological Properties of SANORG 34006, a Potent and Long-Acting Synthetic Pentasaccharide Blood, June 1, 1998; 91(11): 4197 - 4205. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J.M. Herbert, J.P. Herault, A. Bernat, R.G.M. van Amsterdam, G.M.T. Vogel, J.C. Lormeau, M. Petitou, and D.G. Meuleman Biochemical and Pharmacological Properties of SANORG 32701: Comparison With the `Synthetic Pentasaccharide' (SR 90107/ORG 31540) and Standard Heparin Circ. Res., September 1, 1996; 79(3): 590 - 600. [Abstract] [Full Text]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||